You know its strange - We work with an incredible number of decimal facts (financial data), but we find that this has zero impact on our SQLite usage because we have policy where extremely complex business entities that are prone to rapid change are serialized to JSON and stored as BLOB columns. Within this JSON serialized data is where virtually all of our decimal values reside. Obviously, its not a solution for any arbitrary case, but you can usually find workarounds in interesting places further up the stack. We use only 3 data types in SQLite - BINARY, TEXT and INTEGER.
There is no BINARY type in SQLite. (However, the declared type mostly does not restrict the data that column may store. All columns can store blobs and nulls (unless NOT NULL is specified). Columns declared CHAR, VARCHAR, CLOB, CHARACTER, or TEXT can store only text, blob, and null; trying to store a number will convert it to text. Columns declared REAL, FLOAT, or DOUBLE cannot store integers; they are converted into floating numbers; if you try to store text representing a number, it is converted to that number. Columns declared INT, INTEGER, NUMERIC, or an unrecognized type (such as BINARY) can store any value, but text representing a number is converted to a number, and floating numbers which can be losslessly represented as integers are converted to integers. Columns declared BLOB or with no type specified can store any value.)
